Project Summary/Abstract: The goal of the proposed research is to investigate the impact of screening, brief intervention, and referral to treatment (SBIRT) for alcohol and drug use on subsequent health care utilization. We propose further analysis of a unique existing data set, which includes data generated by Wisconsin?s SAMHSA-funded primary care SBIRT program, linked to Medicaid claims data for 16,789 Medicaid recipients. The Wisconsin Initiative to Promote Healthy Lifestyles (WIPHL) SBIRT program provided services in diverse primary care sites, using paraprofessionals in the universal delivery of SBIRT. This innovation is central given the time and financial constraints of physician-led SBIRT. If the analysis yields promising results, this study will serve as preliminary data for an R01 proposal. The study design uses an SBIRT treatment group of 7,367 adult Medicaid beneficiaries (ages 18-64) from 22 primary care sites, and a comparison group of 9,422 sex-matched beneficiaries from 41 comparable non- SBIRT clinic sites. Primary utilization outcomes are: 1) outpatient visits (days); 2) inpatient length of stay (days); inpatient admissions; 4) emergency department admissions; and 5) utilization of substance abuse and mental health specialty care services, all during one year before and two years after the SBIRT index date. Associated costs will also be estimated based on reimbursement rates calculated from Medicaid fee-for- service claims. To assess the primary aim of estimating overall impact on health care utilization, a hierarchical (individuals clustered within clinics) difference-in-differences approach will be used to assess change in utilization between the two groups. A second aim of analysis will investigate type-2 diabetes as a potential moderator of SBIRT?s impact on health care utilization. Risky alcohol use among diabetics can significantly affect type-2 diabetes- related health and health care visits, increasing the cost of managing diabetes. We hypothesize that SBIRT services for individuals with diabetes will significantly reduce the use of high-cost inpatient and emergency department care, thereby decreasing overall health care cost due to better diabetes management and control. A third aim will investigate the effectiveness of screening only compared to screening plus brief intervention among individuals who screened positive for binge drinking. This proposed research using existing data will lead to future rigorous research on the effectiveness of paraprofessional-administered SBIRT, and is likely to inform the decisions of Medicaid systems and other health insurers regarding SBIRT coverage and services provided by paraprofessionals. The study will expand the evidence base regarding SBIRT and, assuming positive cost offset findings, could significantly impact the uptake of SBIRT paraprofessional-led services. Studying diabetes as a moderator could influence diabetes care teams and diabetes support programs to include alcohol screening and brief intervention as a primary strategy for improving self-management behaviors. The study will also elucidate the added effectiveness of brief intervention compared to screening alone.